Topic covering feature updates/bugfixes.

 

Sep. 29 2017 - HDMI hotplug script/udev added for Mainline.  Switching monitors no longer results in a necessary reboot.

Nov. 22 2017 - Dev updated to 4.14 with Libretech patches

Nov. 23 2017 - Next updated to 4.14.y with Libretech patches  <- wait for nightly, not on download page Active download

Dec. 18 2017 - Patch uploaded to fix Ethernet instability: commit

Feb. 04 2018 - Ethernet LED's fixed in repo. commit

July. 08 2018 - Next is kernel 4.17, HDMI audio fixed

 

[bug] Dec. 20 2017 - Init 0 results in reboot

[bug] July 8 2018 - Kernel 4.17 cpufreq throwing errors setting clock speeds, even after BayLibre patch. (board was initially unbootable, patch fixed that)

Le Potato Dev updated with 4.14 Libretech patchset, thanks to @Neil Armstrong . 

 

Anyone willing please build/test, the intention is this become the "next" image.

Good, please tell me if an issue appears with this patchset ! I cleaned it up because it was insane !

Just FYI, mainline U-boot is ongoing, I hope Potato will show up in next version, meanwhile the following tree can be used :

https://github.com/BayLibre/u-boot/tree/u-boot/libretech-cc

3 hours ago, Neil Armstrong said:

Just FYI, mainline U-boot is ongoing,

I saw that, and was going to play with it next.  Thanks again!

Beware using my 4.14 patchset to have it booting, I hope it will work with 4.15 mainline, but for sure it will for 4.16

If you stick with my 4.14 patchset it should work

@Igor I think we can push 4.14 to NEXT, I have it ready to push, but there is the disabled RT patch that I haven't looked at.  (I'd forgotten all about it honestly)  Leave it for review later?  And can I push the changes?

 

Changes:

 

 -NEXT to 4.14.y (LTS?)

 -Patchset as discussed above.

 

 _                       _        _
| |    ___   _ __   ___ | |_ __ _| |_ ___
| |   / _ \ | '_ \ / _ \| __/ _` | __/ _ \
| |__|  __/ | |_) | (_) | || (_| | || (_) |
|_____\___| | .__/ \___/ \__\__,_|\__\___/
            |_|

Welcome to ARMBIAN 5.35 user-built Debian GNU/Linux 9 (stretch) 4.14.1-meson64
System load:   1.48 0.38 0.13   Up time:       0 min
Memory usage:  3 % of 1850MB    IP:            10.0.0.101
Usage of /:    9% of 29G

New to Armbian? Check the documentation first: https://docs.armbian.com

This kernel/patchset likewise results in a workable NanoPi K2 (minus wifi at the moment), can be tested.

Push it and ignore that RT patch. Its probably not working at 4.14 and we need another one. It's low priority at the moment.

https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git/diff/arch/arm64/boot/dts/amlogic/meson-gxl.dtsi?id=v4.14.3&id2=v4.14.2

 

4.14.3 has the memory fix built in for s905x

@Neil Armstrong I'm looking for others to verify this, but on my setup the left and right audio channels are swapped over HDMI.  I verified it wasn't my monitor/headphone configuration by simply changing the monitors selected input to my main computer, where the channels were correct.

 

I am also seeing a random MAC address, is there an official fix for that, or is it a workaround situation?

@TonyMac32btw do you have working drm drivers( I don't have /dev/dri )? I can't make it to work on Odroid C2.

Random mac adress fix will be fine too

 

@Neil Armstrong fixed  mali drivers on 4.14 kernel. So I hope We will have working wayland

The C2 is a gxbb board, it would be better discussed in the K2 topic.  

 

Me being a smartass aside, I will take a look.  For Le Potato I'm going to be testing out the Ethernet patches to hopefully cure the network dropping out.

 

*AT YOUR OWN RISK* You may be able to test the nanopi K2 image on the C2, they are both GXBB, I'd recommend comparing the device trees to look for any issues.

 

The network issue appears resolved as of the gxl patches submitted a day or two ago.  I have enabled usb sound (modules), camera and cifs and ecryptfs.  I have one issue the hdmi sound works sometimes and not others,  The only messages that show up that might be related in syslog are:

Dec 21 08:08:39 localhost pulseaudio[12024]: [alsa-sink-meson-i2s-dai.0.auto-i2s-hifi i2s-hifi-0] alsa-sink.c: ALSA woke us up to write new data to the device, but there was actually nothing to write.
Dec 21 08:08:39 localhost pulseaudio[12024]: [alsa-sink-meson-i2s-dai.0.auto-i2s-hifi i2s-hifi-0] alsa-sink.c: Most likely this is a bug in the ALSA driver '(null)'. Please report this issue to the ALSA developers.
Dec 21 08:08:39 localhost pulseaudio[12024]: [alsa-sink-meson-i2s-dai.0.auto-i2s-hifi i2s-hifi-0] alsa-sink.c: We were woken up with POLLOUT set -- however a subsequent snd_pcm_avail() returned 0 or another value < min_avail.
Dec 21 09:40:38 localhost pulseaudio[12024]: [alsa-source-USB Audio] alsa-source.c: ALSA woke us up to read new data from the device, but there was actually nothing to read.
Dec 21 09:40:38 localhost pulseaudio[12024]: [alsa-source-USB Audio] alsa-source.c: Most likely this is a bug in the ALSA driver 'snd_usb_audio'. Please report this issue to the ALSA developers.
Dec 21 09:40:38 localhost pulseaudio[12024]: [alsa-source-USB Audio] alsa-source.c: We were woken up with POLLIN set -- however a subsequent snd_pcm_avail() returned 0 or another value < min_avail.

 

Is this unique to me or have others seen hdmi stop working?

Edited December 23, 20178 yr by Tido
addded code box

22 minutes ago, reblair said:

Is this unique to me or have others seen hdmi stop working?

 

I have not seen it, however it's not impossible, the patches themselves refer to the sound driver as WIP, like I noted in the first post (which I update periodically), I am getting a left/right channel swap over HDMI.  Those errors do look like that is something to do with ALSA itself or the USB line in (I'm assuming since it's a source), since the error is mirrored across both devices.

 

I'm glad to see the networking appears to be resolved, so far I have not had any issues either.

On 12/14/2017 at 6:14 AM, TonyMac32 said:

@Neil Armstrong I'm looking for others to verify this, but on my setup the left and right audio channels are swapped over HDMI.  I verified it wasn't my monitor/headphone configuration by simply changing the monitors selected input to my main computer, where the channels were correct.

 

Hum, I'd have check this out but I would not be surprised if the channels were swapped. Audio is still a "work in progress" at the moment. I'll try to have look at this later on. Thanks for reporting the problem

With respect to the audio missing, it is true there is a mic equipped camera plugged in and oddly the issue must be intermittent because a reboot fixed it and it hasn't recurred after a day of running and occasional audio use (the camera still connected). 

Ethernet Patch definitely made my potato solid.  I'm running fresh image now, with consul, nomad, docker and Jira running in a container..  def much faster than on my Opi2e.

 

Thanks for the patches!

 

Linux lepotato 4.14.8-meson64 #2 SMP PREEMPT Sun Dec 24 21:03:07 EST 2017 aarch64 aarch64 aarch64 GNU/Linux

 

Could you help me where can I reach that Ethernet patch, how can I install it?

Thanks!

apt update

apt upgrade

 

Use a wireless dongle if available to make it easier.

@Da Xue,  Just lately (in December) I downloaded the 2,2GB image from loverpi (aml-s905x-cc-ubuntu-xenial-linux-libretech-4.14.5-gd9ecccb8d-dirty-desktop-2017-12-17.img.zip) .

I read about the Keyboard bug and boom, it caught me as well. So I was looking here for an update: http://share.loverpi.com/board/libre-computer-project/libre-computer-board-aml-s905x-cc/image/ubuntu/   and yes there is a newer version than I have. So I did: apt-get update && apt-get upgrade  - no updates available.

 

How do you deliver updates to your images ?

 

22 hours ago, Tido said:

How do you deliver updates to your images ?

1

 

Well, only with Armbian (beta) 

@Tido, Armbian is way better at this than we are. Right now apt-get update only updates the debian packages and not the kernel, u-boot, and any other image fixes. If you are using standard Ubuntu, Armbian is the way to go.

 

We have purchased a domain which we will package some u-boots and kernels but that won't happen until a non-preview release. Hence, why our images are "preview" releases rather formal releases. We anticipate the formal releases to happen around the time 18.04 comes out. Our images will have a different target than Armbian since we will be including some specific support for ecosystem components like FPGAs and HATs and our target markets. Armbian will be the more generic distro for these boards.

15 hours ago, Da Xue said:

that won't happen until a non-preview release

So you will start testing when you have a release available, but you might face problems then, because of no testing.
Why not start doing testing already and  name it like:  "Alpha or Beta  upgrade path - it might break your system" and the user has to activate it by themselv like in armbian ?

@Tido, I found the ethernet LED's and a couple of other adjustments that weren't part of the 4.14 patchset.  Quick test then I'll push it.

brilliant, let me know if I can do some tests for you.

Thanks @Tido, this was a simple device tree tweak, so lack of errors = AOK for now.  I noticed the gpio names changed with the pin assignment fix, it doesn't call them gpio0 and gpio10 anymore, just gpio 0 and 1.  Only the activity LED seems to come on (amber) for me, but given my board being a historical artifact that could be the cause.  ;-)

 

Did some housekeeping on the patches, the BayLibre work is filtering back to 4.14 so the patches for bugs are dropping, at this point the patches are primarily for WIP items.

will this change show up in the Beta ?

I would assume, if indeed all packages are rebuilt nightly, that this would pop up in beta tomorrow, it was too late last night to hit today I think.  To be honest I haven't spent time understanding the build frequencies/etc, something I should probably figure out.

On 4.2.2018 at 6:45 PM, TonyMac32 said:

Only the activity LED seems to come on (amber)

I can confirm the green = Link is dead.  Ethernet 10/100M (ETH MDI diff x 2)

 

Ethernet-Port on the schematic is on sheet 7,

GPIOZ_14 (SPDIF_IN // ETH_LINK_LED)  -  P20 -
GPIOZ_15 (PWM_C // ETH_ACT_LED)  -  P19 -

 

not really, I think they forgot to connect the green LED ,  the part 5J1 is the Single Port Connector with Magnetics Module and LED and refers to:

HY911105H  https://aw-som.com/product_info.php?products_id=54

HR911105A  http://www.datasheetmeta.com/search.php?q=HR911105A

 

@Da Xue, can you check whether the green LED (pin 9+10) is only missing in the schematics or did you forget to connect it ?

 

Is there a way to set the MAC address of the ethernet? I can't find an option for modprobe for this driver, unless I'm looking in the wrong place.  I'm trying to just use IPv6 to ssh to this board but annoyingly the MAC keeps changing and thus so does the IPv6 addr. 

 

I'm sure this is the wrong option line I'm using, any guidance on what the proper way to do this is?

 

chris@lepotato:~$ cat /etc/modprobe.d/dwmac_meson8b.conf 
options rtw_initmac=00:e0:4c:f5:16:d8

 

EDIT: Ignore me. I can be an idiot sometimes. Never had to set an interface MAC manually, but if anyone else is looking, just look in /etc/network/interfaces for the example.